惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

WordPress大学
WordPress大学
博客园 - 司徒正美
宝玉的分享
宝玉的分享
阮一峰的网络日志
阮一峰的网络日志
The Cloudflare Blog
月光博客
月光博客
博客园 - 【当耐特】
让小产品的独立变现更简单 - ezindie.com
让小产品的独立变现更简单 - ezindie.com
奇客Solidot–传递最新科技情报
奇客Solidot–传递最新科技情报
博客园 - 三生石上(FineUI控件)
博客园 - 聂微东
小众软件
小众软件
Hugging Face - Blog
Hugging Face - Blog
酷 壳 – CoolShell
酷 壳 – CoolShell
Apple Machine Learning Research
Apple Machine Learning Research
V
V2EX
Jina AI
Jina AI
钛媒体:引领未来商业与生活新知
钛媒体:引领未来商业与生活新知
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
罗磊的独立博客
雷峰网
雷峰网
博客园 - 叶小钗
量子位
IT之家
IT之家

Recent Commits to openclaw:main

test: merge chat side-result checks · openclaw/openclaw@ddd2c2a test: merge cron history checks · openclaw/openclaw@f7eb746 test: merge responsive navigation shell checks · openclaw/openclaw@c2e4b47 docs(changelog): add codex oauth fixes · openclaw/openclaw@628e6cd test: merge navigation routing cases · openclaw/openclaw@5d8cecb Tests: mock channel registry bundled fallback · openclaw/openclaw@2b08233 Secrets: avoid broad web search discovery for single plugin config · openclaw/openclaw@a464f59 test: merge config view browser checks · openclaw/openclaw@20cf511 fix(status): align oauth health with runtime · openclaw/openclaw@eed7116 feat: add macOS screen snapshots for monitor preview (#67954) thanks … · openclaw/openclaw@f377db1 fix: report shared auth scopes in hello-ok (#67810) thanks @BunsDev · openclaw/openclaw@0b6c39b Auto-reply: avoid eager bundled route fallback · openclaw/openclaw@3ea1bf4 Tests: narrow session binding contract setup · openclaw/openclaw@54e4e16 fix(macOS): enable undo/redo in webchat composer text input (#34962) · openclaw/openclaw@00951dc Tests: speed up channel setup promotion · openclaw/openclaw@82b529a Docs: refresh agent instructions · openclaw/openclaw@5775fe2 fix(auth): serialize OAuth refresh across agents to fix #26322 (#67876) · openclaw/openclaw@8e79080 test: allow ollama public surface boundary test · openclaw/openclaw@7d4f1a6 Docs: add test performance guardrails · openclaw/openclaw@89706d3 Tests: restore context-engine usage proof · openclaw/openclaw@e4c4f95 Tests: slim context engine runtime coverage · openclaw/openclaw@74c198f ci: retry failed custom checkouts · openclaw/openclaw@0ee5baf test: trim duplicate provider auth onboarding cases · openclaw/openclaw@1ffc02e matrix: fix sessions_spawn --thread subagent session spawning (#67643) · openclaw/openclaw@1ce2596 test: reduce auth choice fixture churn · openclaw/openclaw@857b9cd test: mock health status config boundaries · openclaw/openclaw@9d5ab4a test: mock onboard config io boundary · openclaw/openclaw@299694d test: mock legacy state plugin boundaries · openclaw/openclaw@2713089 test: mock channel install boundaries · openclaw/openclaw@b945248 test: mock doctor preview channel boundaries · openclaw/openclaw@b1a3ad4
fix: strip response-only reasoning fields from OpenAI Com...
sliverp · 2026-05-15 · via Recent Commits to openclaw:main

@@ -5413,3 +5413,71 @@ describe("openai transport stream", () => {

54135413

).rejects.toThrow("Exceeded tool-call argument buffer limit");

54145414

});

54155415

});

5416+5417+

describe("buildOpenAICompletionsParams strips response-only reasoning fields", () => {

5418+

// OpenRouter and other OpenAI-Completions providers echo `reasoning_details`

5419+

// (array) and `reasoning_content` (string) in response choices to expose the

5420+

// model's chain of thought. If those fields leak back into a follow-up

5421+

// request, OpenRouter rejects the call with HTTP 500 ("Internal Server

5422+

// Error"). buildOpenAICompletionsParams must scrub them before the wire.

5423+

// Repro recipe (verified against live OpenRouter):

5424+

// POST /chat/completions with messages[*].reasoning_details: "..." => 500

5425+

// Same body without that key => 200

5426+5427+

const baseModel = {

5428+

id: "deepseek/deepseek-v4-flash",

5429+

name: "DeepSeek v4 Flash",

5430+

api: "openai-completions",

5431+

provider: "openrouter",

5432+

baseUrl: "https://openrouter.ai/api/v1",

5433+

reasoning: true,

5434+

input: ["text"],

5435+

cost: { input: 0, output: 0, cacheRead: 0, cacheWrite: 0 },

5436+

contextWindow: 128_000,

5437+

maxTokens: 8192,

5438+

} satisfies Model<"openai-completions">;

5439+5440+

function getAssistantMessage(params: { messages: unknown }) {

5441+

expect(Array.isArray(params.messages)).toBe(true);

5442+

const list = params.messages as Array<Record<string, unknown>>;

5443+

const assistant = list.find((m) => m.role === "assistant");

5444+

expect(assistant).toBeDefined();

5445+

return assistant as Record<string, unknown>;

5446+

}

5447+5448+

it("removes reasoning_details, reasoning_content, and reasoning from assistant replay", () => {

5449+

const params = buildOpenAICompletionsParams(

5450+

baseModel,

5451+

{

5452+

systemPrompt: "system",

5453+

messages: [

5454+

{ role: "user", content: "你好" },

5455+

{

5456+

role: "assistant",

5457+

provider: "openrouter",

5458+

api: "openai-completions",

5459+

model: "deepseek/deepseek-v4-flash",

5460+

stopReason: "stop",

5461+

timestamp: 0,

5462+

content: [

5463+

{

5464+

type: "thinking",

5465+

thinking: "User said hi, I should respond politely.",

5466+

thinkingSignature: "considering",

5467+

},

5468+

{ type: "text", text: "Hello!" },

5469+

],

5470+

},

5471+

{ role: "user", content: "再来一个" },

5472+

],

5473+

tools: [],

5474+

} as never,

5475+

undefined,

5476+

) as { messages: unknown };

5477+5478+

const assistant = getAssistantMessage(params);

5479+

expect(assistant).not.toHaveProperty("reasoning_details");

5480+

expect(assistant).not.toHaveProperty("reasoning_content");

5481+

expect(assistant).not.toHaveProperty("reasoning");

5482+

});

5483+

});